Use fences to exclude livestock from shrublands

How is the evidence assessed?
  • Effectiveness
    51%
  • Certainty
    60%
  • Harms
    10%

Study locations

Key messages

 

  • Two replicated, controlled, randomized studies (one of which was also a before-and-after trial) and one controlled before-and-after trial in the UK found that using fences to exclude livestock increased shrub cover or abundance. Two replicated, controlled, randomized studies in Germany and the UK found that using fences increased shrub biomass or the biomass and height of individual heather plants. Two controlled studies (one of which was a before-and-after study) in Denmark and the UK found that heather presence or cover was higher in fenced areas that in areas that were not fenced. However, one site comparison study in the USA found that using fences led to decreased cover of woody plants. Three replicated, controlled studies (one of which was a before and after study) in the USA and the UK found that fencing either had a mixed effect on shrub cover or did not alter shrub cover.
  • One randomized, replicated, controlled, paired study in the UK found that using fences to exclude livestock did not alter the number of plant species, but did increase vegetation height and biomass. One controlled, before-and-after study in the UK found that fenced areas had lower species richness than unfenced areas.
  • One randomized, replicated, controlled, before-and-after trial in the UK and one site comparison study in the USA found that using fences to exclude livestock led to a decline in grass cover. However, four controlled studies (one of which a before-and-after trial) in the USA, the UK, and Finland found that using fences did not alter cover of grass species. One site comparison study in the USA and one replicated, controlled study in the UK recorded an increase in grass cover.
  • One controlled study in Finland found that using fences to exclude livestock did not alter the abundance of herb species and one site comparison in the USA found no difference in forb cover between fenced and unfenced areas. One replicated, controlled study in the USA found fencing had a mixed effect on herb cover.

About key messages

Key messages provide a descriptive index to studies we have found that test this intervention.

Studies are not directly comparable or of equal value. When making decisions based on this evidence, you should consider factors such as study size, study design, reported metrics and relevance of the study to your situation, rather than simply counting the number of studies that support a particular interpretation.

Supporting evidence from individual studies

  1. A controlled study in 1974–1978 in a heathland in Denmark (Bulow-Olsen 1980) found that using fences to exclude livestock from shrublands increased the presence of common heather Calluna vulgaris but did not affect the presence of wavy-hair grass Deschampsia flexuosa. Five years after fence construction, common heather was present in 70% of fenced plots while it was only present in 13% of unfenced plots. However, the percentage of plots in which wavy-hair grass was present did not differ significantly between fenced (82%) and unfenced areas (81%). In 1974 eighteen 25 m2 areas were fenced to exclude livestock. Every year in 1974–1978 vegetation was recorded in eighteen 1 m2 plots located in the fenced areas and thirty-two 1 m2 plots in the unfenced areas.

    Study and other actions tested
  2. A controlled study in 1989–1995 in upland heathland in northern Finland (Virtanen 1998) found that using fences to exclude herbivores did not alter the abundance five shrub species, two herb species, and two grass species. After six years, the cover of five shrub species was not significantly different in areas that were fenced (0–21%) and areas that were left unfenced (0–12%). The same pattern was true for two of two herb species (fenced: 0–10%; unfenced: 0–9%) and two of two grass species (fenced: 1–42%; unfenced: 1–42%). In 1989 twenty 2,000 cm2 blocks were cut from a natural heath habitat and transplanted to an area with no vegetation. Following this, 10 blocks were fenced and another 10 blocks were left unfenced. Vegetation cover was assessed every year in 1990–1995 using a 625 cm2 plastic sheet with 50 holes drilled in it.

    Study and other actions tested
  3. A replicated, controlled, randomized, before-and-after trial in 1993–1999 in four heathland sites in the UK (Hartley & Mitchell 2005) found that using fences to exclude livestock increased the cover of heather Calluna vulgaris and reduced grass cover. In four of four cases, after six years heather cover in areas that were fenced increased by 5–22% relative to the same areas before fencing, whereas heather cover in areas that were not fenced decreased by 23–31%. In three of four cases, after six years, grass cover declined by 9–27% in areas that were fenced, whereas in areas that were not fenced grass cover increased by 19–30%. In 1993 two 60 m2 plots were fenced and two plots were left unfenced at each site. Nitrogen, phosphorous, and potassium were applied to half of the fenced and unfenced plots. Twelve 1 m2 quadrats were randomly placed in the plots and vegetation cover was estimated by eye every year in 1993-1999.

    Study and other actions tested
  4. A replicated, controlled, randomized study in 2001–2002 in a heathland in Germany (Fottner et al. 2007) found that using fences to exclude sheep increased the biomass of heather Calluna vulgaris and wavy hair grass Deschampsia flexuosa. After 1 year, the biomass of heather was higher in plots that had been fenced (8,270 kg/ha) than in unfenced plots (6,259 kg/ha). Similarly, the biomass of wavy hair-grass was higher in plots that had been fenced (404 kg/ha) than in unfenced plots (303 kg/ha). In 2001 forty 2 m2 plots were randomly located in the heathland. Plots were divided in two with one half fenced and the other left open to grazing sheep. In 2002 plants were harvested from plots, then air dried and weighed to estimate biomass.

    Study and other actions tested
  5. A replicated, controlled, randomized study in 2002–2006 in two degraded moorlands in the UK (Mitchell et al. 2008) found that using fences to exclude livestock increased the abundance of heather Calluna vulgaris plants in one of two sites after three years, but decreased the presence of heather in the other case. At one site heather plant density was higher in areas that were fenced to exclude livestock (38 plants/m2) than in unfenced areas (16 plants/m2), while at the other site there was no significant difference between fenced and unfenced areas. Similarly, at one site heather plants were present in fewer of the fenced plots (39%) than the in unfenced plots (46%), while at the other site there was no significant difference between fenced and unfenced areas. At each site fifty-four 100 m2 plots were established. Half of each plot was fenced with the other half left unfenced. Plant cover was estimated annually, between 2003 and 2006, using nine 1 m2 quadrats in each plot.

    Study and other actions tested
  6. A replicated, controlled, randomized study in 2002–2010 in a former heathland in the UK (Critchley et al. 2013) found that in two of three cases using fences to exclude livestock increased the height and biomass of heather Calluna vulgaris plants. After eight years, fenced areas had taller heather plants (31 cm) than unfenced areas grazed by sheep or both sheep and cattle (11–12 cm), but areas grazed by just cattle did not differ significantly in height (27 cm). Heather plants in fenced areas also had higher biomass (17 g) than those in areas grazed by sheep (2 g) or both sheep and cattle (3 g), but heather plants in areas grazed by only cattle did not have significantly lower biomass (12 g). In 2002, nine 5–7 ha paddocks were selected. Paddocks were grazed with cattle, sheep, or left ungrazed. In each paddock six 10 x 10 m plots were established. These plots were divided into four 4 x 4 m subplots two of which were fenced to reduce grazing and two of which were left unfenced. Vegetation height was recorded in each subplot in May 2010. In May 2010 heather was also cut and weighed to calculate biomass.

    Study and other actions tested
  7. A site comparison study in 1958–2008 in sagebrush scrub in Nevada, USA (Rickart et al. 2013) found that using fences to exclude livestock decreased cover of woody plants and increased cover of grasses, but did not alter forb cover after 50 years. In fenced areas cover of woody plants (10%) was lower than in unfenced areas (38%), but for grass species the opposite was true (fenced: 32%, unfenced: 5%). There was no significant difference in the cover of forb species between fenced (10%) and unfenced areas (6%). In the 1950s part of the site was fenced to exclude livestock. In 2008 vegetation cover was estimated in 10 randomly placed 1 m2 quadrats in both the fenced and unfenced areas.

    Study and other actions tested
  8. A controlled, before-and-after trial in 2003–2012 in a heathland previously affected by fire in Surrey, UK (Groome & Shaw 2015) found that cover of dwarf shrubs increased more over nine years in fenced areas than in unfenced areas, but that cover of purple-moor grass Molinia caerulea remained similar in fenced and unfenced areas. Before fencing, the cover of dwarf shrubs was similar in areas that were subsequently fenced (4–8% cover) and unfenced (5–8% cover). However, after nine years fenced areas had higher cover of dwarf shrubs (38–54%) than unfenced areas (27–36%). Before fencing, cover of purple moor grass was also not significantly different between areas that were subsequently fenced (11–47%) and those that were unfenced (19–45%) and this remained the case nine years after fencing (fenced: 12–63% cover; unfenced: 16–42% cover). The entire heathland was burned by wildfire in 2003. Cattle were introduced to the area in 2005, but three areas were fenced to exclude cattle. In 2003, 2010, and 2012 vegetation cover was recorded in fifty-eight 1 m2 quadrats placed in the fenced area and 116 quadrats in the unfenced area.

    Study and other actions tested
  9. A replicated, controlled, before-and-after trial in 1990-1996 in two moorland sites in Derbyshire, UK (Welch 1998) found that after using fences to exclude sheep the cover of crowberry Empetrum nigrum and grass was higher and the cover of mat-grass Nardus stricta was lower in both sites, whereas the effects on cover of bilberry Vaccinium myrtillus and heather Calluna vulgaris were mixed. In both sites and after six years, cover of crowberry and grass was higher and the cover of mat-grass was lower after excluding sheep (crowberry: 25–28%; grass: 64–92%; mat-grass: 6–8%) than before grazing restriction (crowberry: 16%; grass: 55–81%; mat-grass: 13–15%). Additionally, in one of the sites cover of bilberry decreased (from 62% to 45%) and cover of heather increased (from 28% to 39%). Over the same period in plots with year-round grazing, the cover of bilberry declined in one site from 73% to 56% and the cover of grass declined the other site from 81% to 66%. The heights of bilberry and heather increased respectively from 13–16 cm to 18–21cm and from 19–25 cm to 30–33 cm. In each site, sheep were excluded by fencing from two areas of 50 m x 10 m and two nearby areas had year-round grazing. Vegetation height and cover was recorded annually in August.

    Study and other actions tested
  10. A randomized, replicated, paired, controlled study in 1997–2003 in a heathland site in the UK (Denyer et al. 2010) found that fencing to exclude livestock increased vegetation height and biomass, but had a mixed effect on the number of plant species. After six years and in five of six comparisons, vegetation was taller in fenced areas (32–137 cm) than in unfenced areas (4–16 cm). Plant biomass was higher in fenced areas (1352–2832 g/m2) than in unfenced areas (751 g/m2). In one of two comparisons the number of plant species in fenced areas was lower (9 species) than in unfenced areas (17 species) but in one of two comparisons there was no significant difference in the number of plant species (fenced: 16 species, unfenced 17 species). The site was grazed by approximately 20 sheep throughout the study. Twenty four 4 x 4 m plots that were fenced to exclude livestock were established at the site, along with 12 unfenced plots. Half of the fenced plots could be accessed by rabbits that were present at the site. All plots were paired. In 2003 vegetation height was measured in each plot using a ruler and vegetation was harvested from subplots adjacent to plots to calculate biomass. Vegetation cover of all plant species was recorded in plots 2003.

    Study and other actions tested
  11. A replicated, controlled study in a heathland site in the UK (Wilkie 2013) found that fencing to exclude livestock had no effect on vegetation composition and did not alter the abundance of bentgrass Agrostis spp., festuca species Festuca spp., rush Juncus spp., moor grass Molinia spp., sedge Carex spp., heather Calluna spp., heath Erica spp., or Myrica spp.. Vegetation in grazed areas has a similar community composition to that found in ungrazed areas (data presented as graphical analysis). There were no significant differences in the abundance of bentgrass, festuca, rush, moor grass, sedge, heather, heath, or Myrica spp. between grazed and ungrazed areas (data not presented). There were 0.12 cattle/ha and 0.08 horses/ha in the grazed part of the site. Four fenced plots were established at the site along with four unfenced plots. Within each plot vegetation cover was estimated.

    Study and other actions tested
  12. A replicated, controlled study in 1950–1996 in sagebrush steppe in Idaho, USA (Bork et al. 1998) found that fencing to reduce grazing had mixed effects on shrub, grass, forb, and herb cover. Cover of shrubs in ungrazed areas was lower (21%) than in areas that were grazed only in spring (27%) and not significantly different to shrub cover in areas that were grazed only in autumn (19%). Grass and forb cover were not significantly different in ungrazed areas (grass: 16%, forb: 10%) and areas that were grazed in spring or autumn only (grass: 15–17%, forb: 7–12%). Herb cover in ungrazed areas was lower (25%) than in areas grazed only in autumn (30%) and not significantly different from herb cover in areas grazed only in the spring (23%). In 1950 six areas were fenced, two of which were subsequently grazed by sheep in spring only, two of which were grazed in autumn only, and two of which were not grazed. Vegetation cover was recorded in June 1995 and 1996 in eight randomly placed 1.75 m2 quadrats in each of the fenced areas.

    Study and other actions tested
Please cite as:

Martin, P.A., Rocha, R., Smith, R.K. & Sutherland, W.J. (2020) Shrubland and Heathland Conservation. Pages 483-525 in: W.J. Sutherland, L.V. Dicks, S.O. Petrovan & R.K. Smith (eds) What Works in Conservation 2020. Open Book Publishers, Cambridge, UK.

Where has this evidence come from?

List of journals searched by synopsis

All the journals searched for all synopses

Shrubland and Heathland Conservation

This Action forms part of the Action Synopsis:

Shrubland and Heathland Conservation
Shrubland and Heathland Conservation

Shrubland and Heathland Conservation - Published 2017

Shrubland and Heathland synopsis

What Works 2021 cover

What Works in Conservation

What Works in Conservation provides expert assessments of the effectiveness of actions, based on summarised evidence, in synopses. Subjects covered so far include amphibians, birds, mammals, forests, peatland and control of freshwater invasive species. More are in progress.

More about What Works in Conservation

Download free PDF or purchase
The Conservation Evidence Journal

The Conservation Evidence Journal

An online, free to publish in, open-access journal publishing results from research and projects that test the effectiveness of conservation actions.

Read the latest volume: Volume 21

Go to the CE Journal

Discover more on our blog

Our blog contains the latest news and updates from the Conservation Evidence team, the Conservation Evidence Journal, and our global partners in evidence-based conservation.


Who uses Conservation Evidence?

Meet some of the evidence champions

Endangered Landscape ProgrammeRed List Champion - Arc Kent Wildlife Trust The Rufford Foundation Mauritian Wildlife Supporting Conservation Leaders
Sustainability Dashboard National Biodiversity Network Frog Life The international journey of Conservation - Oryx Cool Farm Alliance UNEP AWFA Bat Conservation InternationalPeople trust for endangered species Vincet Wildlife Trust